I have a working AI health appand zero users. Day 1 of fixing that
Phase is built. It reads women's hormone lab reports and explains them in plain English for perimenopause. It works end to end — upload, AI extraction, explanations, trend tracking.
Users: zero.
Revenue: zero.
I spent months building and about a week marketing, which is exactly backwards. So starting today I'm doing the thing I avoided: talking to actual women in my target audience, one at a time, and asking about their experience instead of pitching.
I'll post what I learn here weekly, good or bad.
For anyone who's done cold user research, what's your opening line that doesn't feel creepy or salesy?
The self-awareness (months building, a week marketing, "exactly backwards") is the right instinct, and talking to users one at a time is the correct fix. On the opening line:
You're not asking about your product, you're asking about their experience, and the opener has to make that obvious in sentence one or people brace for a pitch. What works: "I'm researching how women navigate perimenopause and hormone testing, not selling anything. Would you be open to sharing your experience for 15 minutes? I'm trying to understand what's actually confusing or frustrating about it." No product mention. The moment you name the app, they shift from "sharing my story" to "evaluating your thing," and you lose the honest data.
The bigger flag, more important than the line: as a man building a perimenopause app, cold-messaging women about their hormones will feel creepy no matter how good the opener, if it comes from you cold. The fix isn't a better script, it's a warmer channel. Go where these conversations already happen (perimenopause subreddits, Facebook groups, menopause communities), participate genuinely first, or find one or two women who'll make warm intros. A trusted referral removes the creepiness the opener can't.
The sharpest move given who you are: partner with someone already trusted in the space, a women's health coach, a nurse, a perimenopause creator. They have the relationship you're cold-building, and it solves the messenger problem your product has structurally.
Who have you talked to so far, and did the coldness come from the opener or the context? That tells you if it's a script fix or a channel fix.
This is the most useful comment I've gotten and the least comfortable, which is probably related.
Direct answer to your question: I haven't talked to anyone yet. Zero. So it isn't a script problem or a channel problem, it's that I've been avoiding the conversation entirely and telling myself I was working on the app. The coldness was hypothetical because I never sent anything.
You're right about the structural piece and I've been half-aware of it and not saying it out loud. A 20 year old man DMing women about their hormone panels is not a message that gets better with better wording. No opener fixes the fact of who's sending it. That's not a script I can optimise.
So I'm taking the channel + messenger fix seriously rather than looking for a cleverer line:
Participating in menopause and perimenopause communities publicly, as myself, for weeks before I ask anything of anyone
No cold DMs about hormones. If a conversation happens it's because someone replied to something useful I said in the open
Actively looking for a women's health coach, nurse practitioner, or perimenopause creator to partner with, exactly as you described, someone who already has the trust I'd be trying to manufacture
The partner idea is the one I keep coming back to. It solves a problem I can't out-write, and it makes the product better anyway, because they know what women actually ask and I'm guessing.
If you know anyone in that world worth approaching, I'd take the intro. And if you have a read on what makes an approach to a practitioner land versus get ignored, I'd take that too, that's now the outreach I actually need to get right.
Calling avoidance "working on the app" is the most common and expensive founder trap there is, and naming it out loud is the whole unlock. Respect for sitting with the uncomfortable version instead of the flattering one.
On approaching a practitioner, what makes it land vs get ignored: don't pitch a partnership, offer to solve a problem they already have. Practitioners here are drowning in the same repetitive work, patients showing up confused about their lab numbers, asking the same questions, needing plain-English explanations they don't have time to give one by one. That's literally your app. So the approach isn't "want to partner on my app," it's "I built a tool that explains hormone panels in plain English, I think it could save you the repeated explaining you do with every perimenopause patient, could I show you and get your read on whether it's accurate and useful?"
That reframe leads with their pain not your ask, positions them as the expert validating you (which flatters appropriately and is true), and makes the first interaction about feedback, not endorsement. Endorsement comes later, unforced. Ask for partnership up front and you're a stranger wanting something. Ask them to tell you where your tool is wrong and you're someone respecting their expertise.
The trap: don't send ten generic messages. Approach three, having actually read their content, referencing the specific thing they teach. Warm, researched, singular beats volume when the ask is this relationship-dependent.
Getting that practitioner-pitch exactly right is the kind of positioning problem I spend my time on, I'm part of the team building Hivemind, an AI strategy copilot that pressure-tests exactly this kind of messaging. If it helps, you can run the outreach draft through it: https://hivemind.myosin.xyz. Happy to help you sharpen it either way.
No direct intro handy, but the warmest path is perimenopause creators on Substack and Instagram who already make explainer content, they've self-selected as people who like translating this stuff for a lay audience, exactly your product's job. Start there.
The reframe on the practitioner pitch is the part I'd have gotten wrong. I was going to lead with "want to partner," which is me wanting something from a stranger. Leading with "I built something that might save you the repeated explaining you do with every perimenopause patient, can you tell me where it's wrong" flips it entirely, they're the expert, I'm asking to be corrected, and the endorsement (if it comes) comes later and unforced. That's a much better first contact.
The "approach three, having actually read their content" point is the discipline I need. My instinct is volume, and volume is exactly what got me filtered on Reddit and Facebook. Three warm and researched over ten generic.
And the Substack/Instagram explainer-creator angle is the unlock I didn't see. You're right that they've already self selected as people who like doing my product's job for a lay audience, so they'd immediately know if my explanations are accurate, and their audience is exactly who I'm trying to reach. Starting there this week.
I'll run the outreach draft through Hivemind before I send anything. this is precisely the kind of message I can't afford to get wrong. Thank you, genuinely!!!